The outcomes of patients with heart failure (HF) remain suboptimal. Sodium restriction is the most common non-pharmacologic recommendation for HF patients and is endorsed by most guidelines. Interestingly, this recommendation is based on expert opinion or observational studies. The appropriate sodium intake for HF patients on contemporary therapy remains unknown. Some trials have actually suggested worse outcomes with sodium restriction in HF. High sodium intake can cause fluid retention; however, neurohormonal activation induced by low sodium intake can potentially harm the failing heart also. Therefore, the net impact of sodium restriction on HF outcomes can only be addressed with a well-designed trial testing different levels of sodium restriction. However, this trial cannt rely on patients trying to reduce salt intake, as these attempts would be inconsistent. A full-scal trial based on provision of prepared food with specific sodium content is preferable. Patients with systolic HF (low ejection fraction) discharged after admission for acute HF is a desirable target population because of high mortality and readmission risk. Testing the sodium intake currently recommended for high-risk individuals in the general population (1500 mg/d) vs. 3000 mg/d in patients with HF would be reasonable from an ethical perspective, since >3000 mg/day is unlikely to be beneficial. Critical gaps exist in our knowledge necessary to design such a Phase III clinical trial, including lack of data on: (1) proportion of patients consuming e3000 mg/ sodium after discharge (to be ethically randomized into 1500- vs. 3000-mg/d sodium); (2) outcomes among systolic HF patients stable 2 weeks post discharge (to allow for stabilization of treatment and diet); (3) projected effects of a 1500- vs. a 3000-mg/d sodium diet on outcomes (to power a Phase III trial); (4) on-study retention and compliance with provided food; (5) safety of 1500 and 3000 mg/d sodium, especially related to renal function and blood pressure; (6) optimal visit schedule to track compliance and safety, while maintaining feasibility. Therefore, we propose a pilot study with 2 components. In the registry component, we will evaluate patients with HF and ejection fraction d40% before discharge during admission for acute HF to estimate the proportion of patients willing to participate in a 12-week feeding trial; among those, the proportion consuming e3000 mg/d sodium (based on 24-h urine collection), despite discharge instructions, at the 2-week post-discharge visit. In the randomized component, we will randomize 50 eligible patients into meal plans, prepared in our metabolic kitchen by research nutritionists, with 1500 vs. 3000 mg/d sodium for 12 weeks, followed by an additional 12 weeks of surveillance for clinical and safety events. We will evaluate (1) on-study retention and compliance with prepared food; (2) trends in all-cause mortality, readmissions, and emergency room visits; (3) NT-pro-B-type natriuretic peptide levels; (4) diet palatability and quality of lif; and (5) safety (vital signs and biochemistry). We expect this pilot study to yield information necessary and sufficient to design a full-scale clinical trial that will provide an evidence base fr dietary sodium recommendations in HF.